Rxivist logo

Incorporating Human Movement Data to Improve Epidemiological Estimates for 2019-nCoV

By Zhidong Cao, Qingpeng Zhang, Xin Lu, Dirk Pfeiffer, Lei Wang, Hongbing Song, Tao Pei, Zhongwei Jia, Dajun Zeng

Posted 09 Feb 2020
medRxiv DOI: 10.1101/2020.02.07.20021071

Estimating the key epidemiological features of the novel coronavirus (2019-nCoV) epidemic proves to be challenging, given incompleteness and delays in early data reporting, in particular, the severe under-reporting bias in the epicenter, Wuhan, Hubei Province, China. As a result, the current literature reports widely varying estimates. We developed an alternative geo-stratified debiasing estimation framework by incorporating human mobility with case reporting data in three stratified zones, i.e., Wuhan, Hubei Province excluding Wuhan, and mainland China excluding Hubei. We estimated the latent infection ratio to be around 0.12% (18,556 people) and the basic reproduction number to be 3.24 in Wuhan before the citys lockdown on January 23, 2020. The findings based on this debiasing framework have important implications to prioritization of control and prevention efforts. One Sentence SummaryA geo-stratified debiasing approach incorporating human movement data was developed to improve modeling of the 2019-nCoV epidemic.

Download data

  • Downloaded 7,800 times
  • Download rankings, all-time:
    • Site-wide: 1,025
    • In epidemiology: 129
  • Year to date:
    • Site-wide: 16,926
  • Since beginning of last month:
    • Site-wide: 16,926

Altmetric data


Downloads over time

Distribution of downloads per paper, site-wide


PanLingua

Sign up for the Rxivist weekly newsletter! (Click here for more details.)


News